read_csv - sep : Separator - names : List of column names to use. - header : To use as the column names, and the start of the data. If column names are passed explicitly then the behavior is identical to header=None. - low_memory : It results mixed type inference, so set False to ensure no mixed type. columns=['user_id', 'item_id', 'rating', 'timestamp'] df=pd.read_csv('./u.data',sep='\t', names..